Hugging Face AI Sheets is an open-source, no-code tool for building, enriching, and transforming datasets using AI models. It targets users who want to leverage AI for data manipulation without writing code, offering flexibility through local deployment or Hugging Face Hub integration.

How It Works

AI Sheets utilizes a web-based interface built with Qwik and QwikCity, allowing users to interact with AI models through a spreadsheet-like experience. It integrates with Hugging Face Inference Providers for seamless access to thousands of open-source models. Crucially, it supports custom LLMs that adhere to the OpenAI API specification, enabling users to connect local models (e.g., via Ollama) or other compatible inference endpoints.

Quick Start & Requirements

• Docker: docker run -p 3000:3000 -e HF_TOKEN=your_token_here aisheets/sheets
• Local (pnpm): git clone https://github.com/huggingface/sheets.git && cd sheets && export HF_TOKEN=your_token_here && pnpm install && pnpm dev
• Prerequisites: Hugging Face token (HF_TOKEN) is required for authentication and inference. For custom LLMs, an OpenAI-compatible API endpoint and model name are needed.
• Links: Introduction, Try it out

Highlighted Details

• No-code interface for AI-powered dataset manipulation.
• Supports thousands of Hugging Face Hub models via Inference Providers.
• Extensible to custom LLMs adhering to the OpenAI API specification.
• Text-to-image generation is currently limited to Hugging Face Inference Providers.

Maintenance & Community

The project is maintained by Hugging Face. Community channels and roadmap details are not explicitly provided in the README.

Licensing & Compatibility

The project is released under the Apache-2.0 license, which permits commercial use and linking with closed-source projects.

Limitations & Caveats

Text-to-image generation cannot be customized and will always use Hugging Face Inference Providers. The project relies on environment variables for configuration, which might require careful management in production deployments.